Computational Footprint Research Articles

GapFinisher: A reliable gap filling pipeline for SSPACE-LongRead scaffolder output.

Unknown sequences, or gaps, are present in many published genomes across public databases. Gap filling is an important finishing step in de novo genome assembly, especially in large genomes. The gap filling problem is nontrivial and while there are many computational tools partially solving the problem, several have shortcomings as to the reliability and correctness of the output, i.e. the gap filled draft genome. SSPACE-LongRead is a scaffolding tool that utilizes long reads from multiple third-generation sequencing platforms in finding links between contigs and combining them. The long reads potentially contain sequence information to fill the gaps created in the scaffolding, but SSPACE-LongRead currently lacks this functionality. We present an automated pipeline called gapFinisher to process SSPACE-LongRead output to fill gaps after the scaffolding. gapFinisher is based on the controlled use of a previously published gap filling tool FGAP and works on all standard Linux/UNIX command lines. We compare the performance of gapFinisher against two other published gap filling tools PBJelly and GMcloser. We conclude that gapFinisher can fill gaps in draft genomes quickly and reliably. In addition, the serial design of gapFinisher makes it scale well from prokaryote genomes to larger genomes with no increase in the computational footprint.

A Coupled Tank-in-Series Electrochemical Engineering and Thermal Model for Lithium Sulfur Batteries

Lithium sulfur (LiS) batteries are a promising option for next-generation energy storage due to their high energy density. Applications where weight is especially important like electric flight and electric vehicles have driven the study of this chemistry because lithium ion batteries are approaching their practical limits. For these applications, the demand for packing more energy into a battery can lead to higher resistive batteries. For lithium sulfur batteries in particular, the resistance is seen to increase with higher capacities per cell (1). As the required energy and subsequently battery packs increase, the importance of heat dissipation for safety and performance also grows, which requires an accurate thermal model for design and control. Physics-based models enable a look into the internal states of the battery, which correspond to physical phenomena that can contribute to decreased performance under certain conditions. The mathematical model for lithium sulfur batteries developed by Kumaresan et al (2) was the first one-dimensional physics-based model for lithium sulfur batteries. This discharge model was developed with porous electrode theory and considers transport, kinetics, thermodynamics, and morphological changes. The LiS chemistry includes a reduction of solid sulfur to different order polysulfides and finally a precipitation of Li2S, which leads to complexity due to multiple charge carriers and dynamic electrode morphology. The corresponding physics-based model results in a numerically stiff set of equations with variables evolving across many orders of magnitude. To alleviate the computational footprint and still maintain a high degree of accuracy, we have previously developed a mass-conserving lumped model through volume-averaging, called the Lithium Sulfur Tank-in-Series Model. This model, with average quantities in each region, eliminates the spatial dependence for increased computational efficiency and fewer model parameters that expedites model use in estimation, control, and optimization. In this work, the LiS tank-in-series model is coupled with a thermal model (3). The thermal model includes heat generation due to internal resistance and entropic contributions, and Stroe et al. used the thermal model predictions with a simple empirical model. Coupling the thermal model with a more accurate predictive model allows further insight into battery performance. The model predictions are compared with experimental data from 19.5 Ah pouch cells. Acknowledgments The authors are thankful for financial support from the Battery500 Consortium, BAE Systems, and the Joint Center for Aerospace Technology Innovation. References J. Offer and M. Wild, Lithium Sulfur Batteries, p. 149, John Wiley & Sons, New Jersey (2019).Kumaresan, Y. Mikhaylik, and R. E. White, J. Electrochem. Soc., 155, A576 (2008).I. Stroe, V. Knap, M. Swierczynski, and E. Schaltz, ECS Trans., 77, 467–476 (2017).

SCHEDA: Lightweight euclidean-like heuristics for anomaly detection in periodic time series

Detecting anomalies in time series in real time can be challenging, in particular when anomalies can manifest themselves at different time scales and need to be detected with minimal latency. The need for lightweight real-time algorithms has risen in the context of Cloud computing, where thousands of devices are monitored and deviations from normal behaviour must be detected to prevent incidents. However, this need has yet to be addressed in a way that actually scales to the size of today’s network infrastructures.Typically, time series generated by human activity often exhibit daily and weekly patterns creating long-term dependencies that are difficult to process. In such cases, the euclidean distance between subsequences of the time series, or euclidean anomaly score, can be a very effective tool to achieve good detection within constrained latency; however, this computation has a quadratic complexity and a computational footprint too high for any realistic application.In this paper, we propose SCHEDA (Sampled Causal Heuristics for Euclidean Distance Approximation), a collection of three heuristics designed to approximate the euclidean anomaly score with a low computational footprint in time series with long-term dependencies. Our design goals are a low computational cost, the possibility of real-time operation and the absence of tuning parameters. We benchmark SCHEDA against ARIMA and the euclidean distance and show that in typical monitoring scenarios, it outperforms both at only a fraction of the computational cost.

Identification of transcription factor binding sites using ATAC-seq

Transposase-Accessible Chromatin followed by sequencing (ATAC-seq) is a simple protocol for detection of open chromatin. Computational footprinting, the search for regions with depletion of cleavage events due to transcription factor binding, is poorly understood for ATAC-seq. We propose the first footprinting method considering ATAC-seq protocol artifacts. HINT-ATAC uses a position dependency model to learn the cleavage preferences of the transposase. We observe strand-specific cleavage patterns around transcription factor binding sites, which are determined by local nucleosome architecture. By incorporating all these biases, HINT-ATAC is able to significantly outperform competing methods in the prediction of transcription factor binding sites with footprints.

Iteratively Training Look-Up Tables for Network Quantization

Operating deep neural networks (DNNs) on devices with limited resources requires the reduction of their memory as well as computational footprint. Popular reduction methods are network quantization or pruning, which either reduce the word length of the network parameters or remove weights from the network if they are not needed. In this article, we discuss a general framework for network reduction which we call Look-Up Table Quantization (LUT-Q). For each layer, we learn a value dictionary and an assignment matrix to represent the network weights. We propose a special solver which combines gradient descent and a one-step k-means update to learn both the value dictionaries and assignment matrices iteratively. This method is very flexible: by constraining the value dictionary, many different reduction problems such as non-uniform network quantization, training of multiplierless networks, network pruning, or simultaneous quantization and pruning can be implemented without changing the solver. This flexibility of the LUT-Q method allows us to use the same method to train networks for different hardware capabilities.

TERRAIN AWARE IMAGE CLIPPING FOR REAL-TIME AERIAL MAPPING

Abstract. Many remote sensing applications demand for a fast and efficient way of generating orthophoto maps from raw aerial images. One prerequisite is direct georeferencing, which allows to geolocate aerial images to their geographic position on the earth’s surface. But this is only half the story. When dealing with a large quantity of highly overlapping images, a major challenge is to select the most suitable image parts in order to generate seamless aerial maps of the captured area. This paper proposes a method that quickly determines such an optimal (rectangular) section for each single aerial image, which in turn can be used for generating seamless aerial maps. Its key approach is to clip aerial images depending on their geometric intersections with a terrain elevation model of the captured area, which is why we call it terrain aware image clipping (TAC). The method has a modest computational footprint and is therefore applicable even for rather limited embedded vision systems. It can be applied for both, real-time aerial mapping applications using data links as well as for rapid map generation right after landing without any postprocessing step. Referring to real-time applications, this method also minimizes transmission of redundant image data. The proposed method has already been demonstrated in several search-and-rescue scenarios and real-time mapping applications using a broadband data link and diffent kinds of camera and carrier systems. Moreover, a patent for this technology is pending.

Safety Monitoring and Prognostics for Automatic Aircraft Take-off

The take-off of an aircraft is one of the most dangerous flight phases, as failures and adverse environmental conditions can lead to a catastrophe. Should abnormal events occur during the roll phase, the crew or the flight computer has to make the decision if the take-off can be safely rejected and the aircraft can brake and come to a standstill on the runway, or if the take-off has to be attempted in any case.
 This decision has to be made instantaneously upon the estimate of the current state of the aircraft, using available sensor data under noise and potential failure conditions. In order to do so, at any time during the roll phase, a prediction has to be made, if the aircraft can come to a safe stop within the boundaries of the runway.
 In this paper, we formulate this decision making task as an online prognostics problem and develop a model-based architecture that allows us to perform a probabilistic prediction of the aircraft’s braking distance given the current aircraft state. We are using particle filter and Monte-Carlo based prediction algorithms. Because this task has to be performed in real-time on the on-board flight computer, computational resources are very restricted. We therefore propose several models of increasing fidelity, which have substantially different computational footprints and exhibit different levels of accuracy that can impose severe restrictions on the handling of uncertainties and on the failures that can be modeled.

Fejér-monotone hybrid steepest descent method for affinely constrained and composite convex minimization tasks

ABSTRACTThis paper introduces the Fejér-monotone hybrid steepest descent method (FM-HSDM), a new member to the HSDM family of algorithms, for solving affinely constrained minimization tasks in real Hilbert spaces, where convex smooth and non-smooth losses compose the objective function. FM-HSDM offers sequences of estimates which converge weakly and, under certain hypotheses, strongly to solutions of the task at hand. In contrast to its HSDM's precursors, FM-HSDM enjoys Fejér monotonicity, the step-size parameter stays constant across iterations to promote convergence speed-ups of the sequence of estimates to a minimizer, while only Lipschitzian continuity, and not strong monotonicity, of the derivative of the smooth-loss function is needed to ensure convergence. FM-HSDM utilizes fixed-point theory, variational inequalities and affine-nonexpansive mappings to accommodate affine constraints in a more versatile way than state-of-the-art primal–dual techniques and the alternating direction method of multipliers do. Recursions can be tuned to score low computational footprints, well-suited for large-scale optimization tasks, without compromising convergence guarantees. Results on the rate of convergence to an optimal point are also presented. Finally, numerical tests on synthetic data are used to validate the theoretical findings.

Abstract 581: Determining the Nuclear Transcription Factor Network Controlling Expression of the Cardiac Inotropic Factor S100A1

Down-regulation of cardiac inotropic EF-hand calcium sensor protein S100A1 expression is a hallmark of human heart failure and drives progression and mortality. An AAV-S100A1 gene replacement therapy, which has been shown in preclinical models to rescue chronic heart failure, is in preparation for first clinical trials. However, the nuclear mechanisms that convey S100A1 gene locus silencing in the failing heart have not been determined yet. Given the steady increase of S100A1 expression in the developing heart until young adulthood and its decrease in the aged and failing heart, we developed a novel approach to decipher the mechanisms of S100a1 transcriptional regulation. We performed serial RNAseq analyses in an in vitro cardiac differentiation model of H9c2 cells that displays robust concurrent S100A1 and e.g. sarcoplasmic reticulum calcium ATPase 2 (SERCA2a) mRNA and protein during differentiation. Potential transcription factors (TF) were derived by computational motif analysis from S100A1’s gene locus regulatory elements that delivered more than 200 candidates. Subsequent computational TF footprint binding motive analysis from all differentially expressed transcripts of the serial RNAseq data provided a smaller group of less than 30 putatively TF families in the cH9C2 model that is actively during the in vitro biological process. Additional comparative TF expression analyses with developing rat hearts then enabled subtraction of a final TF group of 10 members for RNA interference screens in the cH9C2 model using S100A1 RNA expression as a read-out. Using a siRNA-mediated TF knock-down screen, we identified a small group of TFs such as zink finger and coiled-coil TFs (e.g. KLFs) and signal transducer and activator of transcription (STATs) TFs with an hitherto unknown ability to independently regulate S100A1 expression. Of note, other inotropic factors such as SERCA2a were found to be positively co-regulated by some of the TFs, such as KLFs, that enhance S100A1 expression. Further analyses were performed to confirm these results. In summary, we present a feasible computational filter and experimental strategy to identify relevant TF networks which we have successfully applied to identify cardiac regulatory networks.

Learning the Regularization Operator for the Optical Flow Problem

The task of displacement estimation for frames of a video sequence is considered. A new convolutional neural network architecture for the optical flow problem is proposed. The method is based on learning the regularization operator for a fast optimization method. The proposed method has low computational complexity and memory footprint at test time. The neural network architecture is based on unrolling iterations of a fast primal-dual method as layers of a convolutional neural network. Iterations of the optimization method are represented as convolutions with filters that are trained on ground truth data by backpropagation. A real-time implementation using graphics processing units is proposed. Experimental results demonstrate an improved quality of the optical flow field as compared to the optimization method based on a fixed regularization operator.

GPU-Accelerated Stochastic Predictive Control of Drinking Water Networks

Despite the proven advantages of scenario-based stochastic model predictive control for the operational control of water networks, its applicability is limited by its considerable computational footprint. In this paper, we fully exploit the structure of these problems and solve them using a proximal gradient algorithm parallelizing the involved operations. The proposed methodology is applied and validated on a case study: the water network of the city of Barcelona.

SDN Based Collaborative Scheme for Mitigation of DDoS Attacks

Software Defined Networking (SDN) has proved itself to be a backbone in the new network design and is quickly becoming an industry standard. The idea of separation of control plane and data plane is the key concept behind SDN. SDN not only allows us to program and monitor our networks but it also helps in mitigating some key network problems. Distributed denial of service (DDoS) attack is among them. In this paper we propose a collaborative DDoS attack mitigation scheme using SDN. We design a secure controller-to-controller (C-to-C) protocol that allows SDN-controllers lying in different autonomous systems (AS) to securely communicate and transfer attack information with each other. This enables efficient notification along the path of an ongoing attack and effective filtering of traffic near the source of attack, thus saving valuable time and network resources. We also introduced three different deployment approaches i.e., linear, central and mesh in our testbed. Based on the experimental results we demonstrate that our SDN based collaborative scheme is fast and reliable in efficiently mitigating DDoS attacks in real time with very small computational footprints.

SmilesDrawer: Parsing and Drawing SMILES-Encoded Molecular Structures Using Client-Side JavaScript.

Here we present SmilesDrawer, a dependency-free JavaScript component capable of both parsing and drawing SMILES-encoded molecular structures client-side, developed to be easily integrated into web projects and to display organic molecules in large numbers and fast succession. SmilesDrawer can draw structurally and stereochemically complex structures such as maitotoxin and C60 without using templates, yet has an exceptionally small computational footprint and low memory usage without the requirement for loading images or any other form of client-server communication, making it easy to integrate even in secure (intranet, firewalled) or offline applications. These features allow the rendering of thousands of molecular structure drawings on a single web page within seconds on a wide range of hardware supporting modern browsers. The source code as well as the most recent build of SmilesDrawer is available on Github ( http://doc.gdb.tools/smilesDrawer/ ). Both yarn and npm packages are also available.

Higher-Fidelity Frugal and Accurate Quantile Estimation Using a Novel Incremental <italic>Discretized</italic> Paradigm

Traditional pattern classification works with the moments of the distributions of the features and involves the estimation of the means and variances. As opposed to this, more recently, research has indicated the power of using the quantiles of the distributions because they are more robust and applicable for non-parametric methods. The estimation of the quantiles is even more pertinent when one is mining data streams. However, the complexity of quantile estimation is much higher than the corresponding estimation of the mean and variance, and this increased complexity is more relevant as the size of the data increases. Clearly, in the context of infinite data streams, a computational and space complexity that is linear in the size of the data is definitely not affordable. In order to alleviate the problem complexity, recently, a very limited number of studies have devised incremental quantile estimators [1] , [2] . Estimators within this class resort to updating the quantile estimates based on the most recent observation(s), and this yields updating schemes with a very small computational footprint–a constant-time (i.e., $O(1)$ ) complexity. In this paper, we pursue this research direction and present an estimator that we refer to as a higher-fidelity frugal [1] quantile estimator. First, it guarantees a substantial advancement of the family of Frugal estimators introduced in [1] . The highlight of the present scheme is that it works in the discretized space, and it is thus a pioneering algorithm within the theory of discretized algorithms. 1 The convergence results that we have proven are based on the theory of stochastic point location [3] , which we advocate as a new tool for solving a large class of online estimation problems. Extensive simulation results show that our estimator outperforms the original Frugal algorithm in terms of both speed and accuracy. 1 The fact that discretized Learning Automata schemes are superior to their continuous counterparts has been clearly demonstrated in the literature. This is the first paper, to our knowledge, that proves the advantages of discretization within the domain of quantile estimation.

EMHP: an accurate automated hole masking algorithm for single-particle cryo-EM image processing.

SummaryThe Electron Microscopy Hole Punch (EMHP) is a streamlined suite of tools for quick assessment, sorting and hole masking of electron micrographs. With recent advances in single-particle electron cryo-microscopy (cryo-EM) data processing allowing for the rapid determination of protein structures using a smaller computational footprint, we saw the need for a fast and simple tool for data pre-processing that could run independent of existing high-performance computing (HPC) infrastructures. EMHP provides a data preprocessing platform in a small package that requires minimal python dependencies to function.Availability and implementation https://www.bitbucket.org/chazbot/emhp Apache 2.0 LicenseSupplementary information Supplementary data are available at Bioinformatics online.

Prediction-based load balancing and resolution tuning for interactive volume raycasting

We present an integrated approach for real-time performance prediction of volume raycasting that we employ for load balancing and sampling resolution tuning. In volume rendering, the usage of acceleration techniques such as empty space skipping and early ray termination, among others, can cause significant variations in rendering performance when users adjust the camera configuration or transfer function. These variations in rendering times may result in unpleasant effects such as jerky motions or abruptly reduced responsiveness during interactive exploration. To avoid those effects, we propose an integrated approach to adapt rendering parameters according to performance needs. We assess performance-relevant data on-the-fly, for which we propose a novel technique to estimate the impact of early ray termination. On the basis of this data, we introduce a hybrid model, to achieve accurate predictions with minimal computational footprint. Our hybrid model incorporates aspects from analytical performance modeling and machine learning, with the goal to combine their respective strengths. We show the applicability of our prediction model for two different use cases: (1) to dynamically steer the sampling density in object and/or image space and (2) to dynamically distribute the workload among several different parallel computing devices. Our approach allows to reliably meet performance requirements such as a user-defined frame rate, even in the case of sudden large changes to the transfer function or the camera orientation.

Indexing quantized ensembles of exemplar-SVMs with rejecting taxonomies

Ensembles of Exemplar-SVMs have been introduced as a framework for Object Detection but have rapidly found a large interest in a wide variety of computer vision applications such as mid-level feature learning, tracking and segmentation. What makes this technique so attractive is the possibility of associating to instance specific classifiers one or more semantic labels that can be transferred at test time. To guarantee its effectiveness though, a large collection of classifiers has to be used. This directly translates in a high computational footprint, which could make the evaluation step prohibitive. To overcome this issue we organize Exemplar-SVMs into a taxonomy, exploiting the joint distribution of Exemplar scores. This permits to index the classifiers at a logarithmic cost, while maintaining the label transfer capabilities of the method almost unaffected. We propose different formulations of the taxonomy in order to maximize the speed gain. In particular we propose a highly efficient Vector Quantized Rejecting Taxonomy to discard unpromising image regions during evaluation, performing computations in a quantized domain. This allow us to obtain ramarkable speed gains, with an improvement up to more than two orders of magnitude. To verify the robustness of our indexing data structure with reference to a standard Exemplar-SVM ensemble, we experiment with the Pascal VOC 2007 benchmark on the Object Detection competition and on a simple segmentation task.

Prune the Convolutional Neural Networks with Sparse Shrink

Nowadays, it is still difficult to adapt Convolutional Neural Network (CNN) based models for deployment on embedded devices. The heavy computation and large memory footprint of CNN models become the main burden in real application. In this paper, we propose a "Sparse Shrink" algorithm to prune an existing CNN model. By analyzing the importance of each channel via sparse reconstruction, the algorithm is able to prune redundant feature maps accordingly. The resulting pruned model thus directly saves computational resource. We have evaluated our algorithm on CIFAR-100. As shown in our experiments, we can reduce 56.77% parameters and 73.84% multiplication in total with only minor decrease in accuracy. These results have demonstrated the effectiveness of our "Sparse Shrink" algorithm.

Binary descriptor-based dense line-scan stereo matching

We present a line-scan stereo system and descriptor-based dense stereo matching for high-performance vision applications. The stochastic binary local descriptor (STABLE) descriptor is a local binary descriptor that builds upon the principles of compressed sensing theory. The most important properties of STABLE are the independence of the descriptor length from the matching window size and the possibility that more than one pair of pixels contributes to a single-descriptor bit. Individual descriptor bits are computed by comparing image intensities over pairs of balanced random subsets of pixels chosen from the whole described area. On a synthetic as well as real-world examples, we demonstrate that STABLE provides competitive or superior performance than other state-of-the-art local binary descriptors in the task of dense stereo matching. The real-world example is derived from line-scan binocular stereo imaging, i.e., two line-scan cameras are observing the same object line and 2-D images are generated due to relative motion. We show that STABLE performs significantly better than the census transform and local binary patterns (LBP) in all considered geometric and radiometric distortion categories to be expected in practical applications of stereo vision. Moreover, we show as well that STABLE provides comparable or better matching quality than the binary robust-independent elementary features descriptor. The low computational complexity and flexible memory footprint make STABLE well suited for most hardware architectures. We present quantitative results based on the Middlebury stereo dataset as well as illustrative results for road surface reconstruction.

Fast nonlinear gravity inversion in spherical coordinates with application to the South American Moho

Estimating the relief of the Moho from gravity data is a computationally intensive non-linear inverse problem. What is more, the modeling must take the Earths curvature into account when the study area is of regional scale or greater. We present a regularized non-linear gravity inversion method that has a low computational footprint and employs a spherical Earth approximation. To achieve this, we combine the highly efficient Botts method with smoothness regularization and a discretization of the anomalous Moho into tesseroids (spherical prisms). The computational efficiency of our method is attained by harnessing the fact that all matrices involved are sparse. The inversion results are controlled by three hyper-parameters: the regularization parameter, the anomalous Moho density-contrast, and the reference Moho depth. We estimate the regularization parameter using the method of hold-out cross-validation. Additionally, we estimate the density-contrast and the reference depth using knowledge of the Moho depth at certain points. We apply the proposed method to estimate the Moho depth for the South American continent using satellite gravity data and seismological data. The final Moho model is in accordance with previous gravity-derived models and seismological data. The misfit to the gravity and seismological data is worse in the Andes and best in oceanic areas, central Brazil and Patagonia, and along the Atlantic coast. Similarly to previous results, the model suggests a thinner crust of 30-35 km under the Andean foreland basins. Discrepancies with the seismological data are greatest in the Guyana Shield, the central Solimoes and Amazonas Basins, the Parana Basin, and the Borborema province. These differences suggest the existence of crustal or mantle density anomalies that were unaccounted for during gravity data processing.